Learning Bayesian network parameters from small data sets: application of Noisy-OR gates

Oniśko, Agnieszka; Drużdżel, Marek J.; Wasyluk, Hanna

doi:10.1016/s0888-613x(01)00039-1

Cited by 250 publications

(147 citation statements)

References 14 publications

Supporting

Mentioning

145

Contrasting

Order By: Relevance

“…Their foremost advantage is a small number of parameters that are sufficient to specify the entire CPT. This leads to a significant reduction of effort in knowledge elicitation from experts [3,5], improves the quality of distributions learned from data [8], and reduces the spacial and temporal complexity of algorithms for Bayesian networks [9,10].…”

Section: Introductionmentioning

confidence: 99%

Knowledge Engineering for Bayesian Networks: How Common Are Noisy-MAX Distributions in Practice?

Zagorecki

Drużdżel

2013

IEEE Trans. Syst. Man Cybern, Syst.

Self Cite

View full text Add to dashboard Cite

Abstract.One problem faced in knowledge engineering for Bayesian networks is the exponential growth of the number of parameters in their conditional probability tables (CPTs). The most common practical solution is application of the noisy-OR (or their generalization, the noisy-MAX) gates, which take advantage of independence of causal interactions and provide a logarithmic reduction of the number of parameters required to specify a CPT. In this paper, we propose an algorithm that fits a noisy-MAX distribution to an existing CPT and we apply it to search for noisy-MAX gates in three existing practical Bayesian networks. We show that noisy-MAX gate provides a surprisingly good fit for as many as 50% of CPTs in these networks. The importance of this finding is that it provides an empirical justification for the use of the noisy-MAX gate as a powerful knowledge engineering tool.

show abstract

Section: Introductionmentioning

confidence: 99%

Knowledge Engineering for Bayesian Networks: How Common Are Noisy-MAX Distributions in Practice?

Zagorecki

Drużdżel

2013

IEEE Trans. Syst. Man Cybern, Syst.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Parameter learning aims to determine the conditional probability distribution of each node under the established BN structure [11,12]. The conditional probability tables can be determined by learning the parameters on the database using a learning algorithm.…”

Section: Bn Learningmentioning

confidence: 99%

Dynamic Risk Assessment in Construction Projects Using Bayesian Networks

Zhang¹,

Wu²,

Skibniewski

et al. 2014

Proceedings of the International Symposium on Automation and Robotics in Construction (IAARC)

View full text Add to dashboard Cite

show abstract

“…The authors of these systems have reported that the sizes of the underlying networks are superlinear in the number of trials [44,45], and that the training time is superlinear in the network size [21,22]. …”

Section: Scalabilitymentioning

confidence: 99%

Selection of patients for clinical trials: an interactive web-based system

Fink

Kokku

Nikiforou

et al. 2004

Artificial Intelligence in Medicine

View full text Add to dashboard Cite

The purpose of a clinical trial is to evaluate a new treatment procedure. When medical researchers conduct a trial, they recruit participants with appropriate health problems and medical histories. To select participants, they analyze medical records of the available patients, which has traditionally been a manual procedure.We describe an expert system that helps to select patients for clinical trials. If the available data are insufficient for choosing patients, the system suggests additional medical tests and finds an ordering of the tests that reduces their total cost. Experiments show that the system can increase the number of selected patients. We also present an interface that enables a medical researcher to add clinical trials and selection criteria without the help of a programmer. The addition of a new trial takes ten to twenty minutes, and novice users learn the functionality of the interface in about an hour.

show abstract

Learning Bayesian network parameters from small data sets: application of Noisy-OR gates

Cited by 250 publications

References 14 publications

Knowledge Engineering for Bayesian Networks: How Common Are Noisy-MAX Distributions in Practice?

Knowledge Engineering for Bayesian Networks: How Common Are Noisy-MAX Distributions in Practice?

Dynamic Risk Assessment in Construction Projects Using Bayesian Networks

Selection of patients for clinical trials: an interactive web-based system

Contact Info

Product

Resources

About